What question did this study set out to answer?

To develop a deep learning framework for automated segmentation and classification of otoscopic images for otitis media diagnosis.

February 5, 2026DiagnosticsOpen Access

Deep Learning-Based Semantic Segmentation and Classification of Otoscopic Images for Otitis Media Diagnosis and Health Promotion

View Full Paper

Ask AI

Bookmark

Authors

AI-based framework classifies otitis media in clinical images, suggesting enhanced diagnostic support.

To develop a deep learning framework for automated segmentation and classification of otoscopic images for otitis media diagnosis.
Implemented semi-supervised learning for recognizing tympanic membrane structures.
Extracted features from segmented regions using convolutional neural network architectures.
Collected and analyzed 607 clinical otoscopic images for training and testing the model.
U-Net model achieved 96.76% pixel accuracy and 71.68% Dice similarity coefficient in segmentation.
Diagnostic accuracy reached 100% for normal ears and AOM, 91.3% for COM on the test set.
Overall classification accuracy of 96.72% indicates high reliability of the AI framework.

Yang et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

February 5, 2026DiagnosticsOpen Access

View Full Paper

Ask AI

Bookmark

Authors

AI-based framework classifies otitis media in clinical images, suggesting enhanced diagnostic support.

To develop a deep learning framework for automated segmentation and classification of otoscopic images for otitis media diagnosis.
Implemented semi-supervised learning for recognizing tympanic membrane structures.
Extracted features from segmented regions using convolutional neural network architectures.
Collected and analyzed 607 clinical otoscopic images for training and testing the model.
U-Net model achieved 96.76% pixel accuracy and 71.68% Dice similarity coefficient in segmentation.
Diagnostic accuracy reached 100% for normal ears and AOM, 91.3% for COM on the test set.
Overall classification accuracy of 96.72% indicates high reliability of the AI framework.

Yang et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark